1、准备磁盘
2、对磁盘进行分区
  1. The operation has completed successfully.
  2. # sgdisk -n 1:0:0 -c 1:"ceph data" -t 1:4fbd7e29-9d25-41b8-afd0-062c0ceff05d /dev/sde
  3. The operation has completed successfully.
  1. # mkfs.xfs -f -i size=2048 /dev/sde1
4、创建osd
  1. # ceph osd create
  2. 11
5、创建osd数据目录
6、挂载分区到osd的数据目录
    7、初始化osd数据目录
    1. # ceph-osd -i 11 --mkfs --mkkey
    8、指定journal分区
    1. 删除自动生成的journal文件
    2. # rm -rf /var/lib/ceph/osd/ceph-11/journal 
    3. 查看要用作journal分区的sde2对应的uuid
    4. # ll /dev/disk/by-partuuid/ | grep sde2
    5. lrwxrwxrwx 1 root root 10 Aug 31 11:58 adc17c4e-4aa8-44dd-8351-f9a1f8895455 -> ../../sde2
    6. # ln -s /dev/disk/by-partuuid/adc17c4e-4aa8-44dd-8351-f9a1f8895455 /var/lib/ceph/osd/ceph-11/journal
    7. 在osd数据目录下写入journal分区的uuid
    8. # echo adc17c4e-4aa8-44dd-8351-f9a1f8895455 > /var/lib/ceph/osd/ceph-11/journal_uuid
    9. 创建journal
    10. # ceph-osd -i 11 --mkjournal
    9、注册osd的认证秘钥
    10、把OSD添加到CRUSH Map
    1. # ceph osd crush add osd.11 0.01459 host=node3
    11、修改数据目录权限
      12、激活osd
      1. # ceph-disk activate --mark-init systemd --mount /dev/sde1